In order to protect the general public from a broad variety of potentially harmful products and chemicals, toxicologic testing for dermal irritancy is routinely performed using laboratory animals. However, the use of animals and their maintenance is becoming increasingly costly, such animal testing suffers from subjectivity and high variability, and the flaw of applying data obtained with animals to humans persists. In addition, there is increasing public opinion that favors reduction or elimination of animal subjects for any cosmetic, industrial, or medical testing. Thus, non-animal, "in vitro" alternative tests are being sought. The purpose of the present proposal is to model dermal irritation and inflammation using normal human epidermal keratinocytes which have been cultured to form a highly differentiated, stratified epidermis-like tissue. By measuring the release of the molecular mediators which start the physiological cascade leading to dermal irritation in vivo , it is hypothesized that this "in vitro" system could predict the dermal irritancy potential of a particular compound. The proposed system allows one to accomplish this in a rapid, cost effective, and reproducible manner without the use of laboratory animals.